KR101088117B1 - Apparatus and method for fast imaging using image sensor - Google Patents

Abstract

A high speed imaging apparatus and method according to at least one embodiment of the present invention includes scanning a sample in a depth direction and capturing a plurality of times to obtain 2D images, and generating a 3D image of the sample using the 2D images. In spite of using a conventional fluorescence microscope, a three-dimensional image of a sample can be obtained as a high-quality, accurate three-dimensional image of a three-dimensional image obtained through a confocal microscope, and a three-dimensional image can be obtained through a confocal microscope. Unlike the case, a three-dimensional image of a sample can be obtained quickly. In particular, the high-speed imaging apparatus and method according to at least one embodiment of the present invention process each of the obtained two-dimensional images in a predetermined manner, and use the processed two-dimensional images to generate a three-dimensional image having excellent contrast ratio and resolution. Because it can be created, three-dimensional images can be obtained at higher speeds by using a computer rather than a mechanical device, without the need for pinholes or precision expensive optical scanning systems, such as laser scanning confocal microscopes.

Description

High speed imaging apparatus and method using an image sensor {Apparatus and method for fast imaging using image sensor}

The present invention relates to imaging devices and methods, and more particularly, to imaging devices and methods for acquiring an image of a sample, such as a cell.

Fluorescence microscopes are most commonly used as optical systems for observing living cells. The fluorescence microscope receives the fluorescent light from the cells stained with the fluorescein dye as an objective lens and acquires the image of the cell by condensing the enlarged cell image with the image sensor. CCD (Charge Coupled Device) is an example of an image sensor. The optical system of the fluorescence microscope is based on a classic Wide Field Microscope, which allows the image of the fluorescent light from the cell to form an image by forming an image on the CCD. There is a problem that it is difficult to obtain an accurate three-dimensional image of the cell. This problem results in reduced contrast and resolution of the obtained three-dimensional image of the cell, especially in the case of fluorescence microscopy using fluorescently stained cells as a sample, which is more severe than the conventional optical microscope. There is a great difficulty in observing the fine structure of the cells accurately.

In order to solve the problem, a so-called 'confocal laser scanning microscope' has emerged. Confocal microscopy overcomes the disadvantages described above by using pinholes as spatial filters to filter out uncondensed light. Therefore, the confocal microscope has a higher contrast ratio and improved resolution than a general wide field microscope, and furthermore, it is possible to obtain an optical sectioned image. The characteristics of this confocal microscope are essential conditions for constructing a three-dimensional image of the cell. Confocal microscopy now reproduces three-dimensional images of cells with high contrast and resolution, making them an indispensable cell imaging system for biotechnology research. However, to make an image with a confocal microscope, a laser beam scanning device for scanning a laser beam to a sample is required, and an optical signal passing through a pinhole to a photo-detector is imaged. You also need a system that can regenerate the data. Therefore, the confocal microscope takes a long time to measure the three-dimensional image has a disadvantage that it is difficult to observe the real-time change of the sample.

The technical problem to be achieved by at least one embodiment of the present invention is that, despite the use of a conventional fluorescence microscope can obtain a three-dimensional image of the sample as a high quality accurate three-dimensional image of the three-dimensional image obtained through a confocal microscope The present invention provides a high-speed imaging device capable of quickly acquiring a three-dimensional image of a sample, unlike when acquiring a three-dimensional image through a confocal microscope.

Another technical method of at least one embodiment of the present invention is to obtain a three-dimensional image of the sample as a high-quality accurate three-dimensional image of the three-dimensional image obtained through a confocal microscope, despite using a conventional fluorescence microscope The present invention provides a high-speed imaging method capable of quickly acquiring a three-dimensional image of a sample, unlike when acquiring a three-dimensional image through a confocal microscope.

Another technical problem to be achieved by at least one embodiment of the present invention is to obtain a three-dimensional image of the sample as a high-quality accurate three-dimensional image of the three-dimensional image obtained through a confocal microscope, despite using a conventional fluorescence microscope Computer-readable recordings that store computer programs for computer-implementing a high-speed imaging method that allows high-speed imaging methods to be acquired quickly, as opposed to acquiring three-dimensional images through a confocal microscope. To provide the medium.

In order to achieve the above technical problem, the high-speed imaging apparatus according to at least one embodiment of the present invention includes an imaging unit for scanning the sample in the depth direction and imaging the plurality of times to obtain two-dimensional images; And an image reconstruction unit which generates a 3D image of the sample using the 2D images.

Here, the imaging unit may acquire the two-dimensional images by moving the sample in the depth direction and photographing the sample a plurality of times.

Here, the imaging unit may acquire the two-dimensional images by moving the imaging device in the imaging unit in a depth direction and photographing the sample a plurality of times.

Here, the imaging unit may acquire the two-dimensional images by moving the tube lens in the imaging unit in the depth direction and photographing the sample a plurality of times.

Here, the imaging unit may acquire the two-dimensional images by moving the objective lens in the imaging unit in the depth direction and photographing the sample a plurality of times.

Here, the image reconstruction unit may process each of the 2D images according to a preset method and generate the 3D image in consideration of the processed results.

In this case, the image reconstruction unit may process the 2D images such that the intensity of each of the 2D images is squared and generate the 3D image in consideration of the processed results.

In this case, the image reconstruction unit may generate the 3D image by considering the 2D images differentiated and differentiated each of the 2D images a plurality of times.

In this case, the image reconstruction unit may filter the low frequency components of each of the 2D images, and generate the 3D image in consideration of the filtered 2D images.

In order to achieve the above technical problem, a high-speed imaging method according to at least one embodiment of the present invention comprises the steps of: (a) scanning a sample in the depth direction and obtaining a plurality of times to obtain two-dimensional images; And (b) generating a 3D image of the sample using the 2D images.

Here, in step (a), the two-dimensional images may be obtained by moving the sample in the depth direction and photographing the sample a plurality of times.

Here, in step (a), the two-dimensional images may be obtained by moving the imaging device in the imaging unit in a depth direction and photographing the sample a plurality of times.

Here, in step (a), the two-dimensional images may be obtained by moving the tube lens in the imaging unit in the depth direction and photographing the sample a plurality of times.

Here, in step (a), the objective lens in the imaging unit may be moved in the depth direction, and the two-dimensional images may be obtained by photographing the sample a plurality of times.

Here, in step (b), each of the 2D images may be processed according to a preset method and the 3D image may be generated in consideration of the processed results.

In this case, step (b) may process the 2D images such that the intensity of each of the 2D images is squared and generate the 3D image in consideration of the processed results.

In this case, step (b) may generate the 3D image by considering the 2D images differentiated and differentiated each of the 2D images a plurality of times.

In this case, step (b) may filter the low frequency components of each of the 2D images and generate the 3D image by considering the filtered 2D images.

In order to achieve the above object, the computer-readable recording medium according to at least one embodiment of the present invention comprises the steps of scanning a sample in the depth direction and obtaining two-dimensional images by imaging a plurality of times; And a computer program for causing the computer to perform the step of generating a 3D image of the sample using the 2D images.

A high speed imaging apparatus and method according to at least one embodiment of the present invention includes scanning a sample in a depth direction and capturing a plurality of times to obtain 2D images, and generating a 3D image of the sample using the 2D images. In spite of using a conventional fluorescence microscope, a three-dimensional image of a sample can be obtained as a high-quality, accurate three-dimensional image of a three-dimensional image obtained through a confocal microscope, and a three-dimensional image can be obtained through a confocal microscope. Unlike the case, a three-dimensional image of a sample can be obtained quickly. In particular, the high-speed imaging apparatus and method according to at least one embodiment of the present invention process each of the obtained two-dimensional images in a predetermined manner, and use the processed two-dimensional images to generate a three-dimensional image having excellent contrast ratio and resolution. Because it can be created, three-dimensional images can be obtained at higher speeds by using a computer rather than a mechanical device, without the need for pinholes or precision expensive optical scanning systems, such as laser scanning confocal microscopes.

In order to fully understand the present invention, the operational advantages of the present invention, and the objects achieved by the practice of the present invention, reference should be made to the accompanying drawings that illustrate preferred embodiments of the present invention and the accompanying drawings.

Hereinafter, a high speed imaging apparatus and method according to at least one embodiment of the present invention will be described with reference to the accompanying drawings.

1 is a block diagram illustrating a high speed imaging apparatus according to at least one embodiment of the present invention, FIG. 2 is a reference diagram of a system representing a high speed imaging apparatus according to a first embodiment of the present invention, and FIG. A reference diagram of a system for expressing a high speed imaging apparatus according to a second embodiment of the present invention. In the present specification, the cell is an example of a sample.

The high speed imaging apparatus according to at least one embodiment of the present invention may be implemented based on a fluorescence microscope as shown in FIG. 2 or 3. Detailed description thereof will be described later with reference to FIGS. 2 and 3. First, the present invention will be described with reference to FIG. 1 as follows.

The imaging unit 110 scans a sample in a depth (so-called 'z') direction and captures a plurality of times to obtain a plurality of two-dimensional images. To this end, the imaging unit 110 may not only be configured with a device that is purely responsible for a photographing function, but may also perform a scanning function. Herein, the scanning function means at least one of a sample and 'a component of at least a part of the photographing function responsible device of the image capturing unit 110' so that the image capturing unit 110 scans the sample in a depth direction and captures the sample a plurality of times. Means a function of moving in the depth direction, and such a scanning function is one of functions of the imaging unit 110 in the present specification. At this time, examples of the components of the photographing function responsible device of the imaging unit 110 include an imaging device, a tube lens, an objective lens, and the like. Details will be described later with reference to FIGS. 2 and 3.

The image reconstructor 120 generates (reconstructs) a three-dimensional image of the sample by using the two-dimensional images acquired by the imaging unit 110. In detail, the image reconstruction unit 120 may process each of the 2D images according to a preset method and generate the 3D image in consideration of the processed results. For example, the image reconstructor 120 may process the 2D images such that the intensity of each of the 2D images is squared (= square) and generate the 3D image in consideration of the processed 2D images. A 3D image may be generated by differentiating each of the 2D images a plurality of times and deriving the differentiated 2D images, or filtering the low frequency components of each of the 2D images and considering the filtered 2D images. You can also create an image.

Hereinafter, the image capturing unit 110 and the image reconstructing unit 120 described above will be described in more detail with reference to FIGS. 2 and 3. For reference, in FIGS. 2 and 3, the sample stage 212 to the image sensor 224 mean components of the imaging unit 110, and the computer 226 is an implementation example of the image reconstruction unit 120.

Figure 2 shows a fluorescence microscope system using a typical single lens. When the excitation light 210 is irradiated onto a sample stained with a fluorescent dye placed on the sample stage 212 (where the sample stage 212 means a place where the sample is placed), a fluorophore attached to the sample Fluorescence light (second) is emitted from the secondary light, the objective lens 214 collects the emitted fluorescent light and serves to create a two-dimensional image in the image sensor 224. The bandpass filter 220 prevents the excitation light 210 from entering the image sensor 224, and only fluorescent light of a wavelength longer than the wavelength of the excitation light emitted from the phosphor attached to the sample may enter the image sensor 224. Pass it through. Since only one cross section of the sample is imaged on the plane of the image sensor 224 by the objective lens 214, the fluorescent light emitted from the other cross section of the sample is blurred and distributed over a wide area of the image sensor 224. In the imaging system of FIG. 2, the distance between the cross-section of the sample to be imaged and the objective lens 214 is Z ₁ , the distance between the objective lens 214 and the image sensor 224 is Z ₂ , and the focal length of the objective lens 214 is determined. Suppose f satisfies the lens formula as shown in equation (1).

At this time, the magnification M between the two-dimensional image size and the actual sample cross-sectional size measured by the image sensor 224 is expressed by Equation 2 below.

The imaging system of FIG. 3 shows a so-called 'infinity-corrected microscope system'. An example using the epi-illumination excitation method irradiated from the same side as the direction in which the excitation light 210 for exciting the phosphor is observed is shown. In this case, unlike in FIG. 2, the excitation light 210 is reflected by the dichroic mirror 218 at 45 ° and then irradiated from the rear portion of the objective lens 214. The dichroic mirror 218 reflects the excitation light 210 and passes the fluorescent light 216 to filter out the reflected or scattered excitation light from the sample and prevent it from going to the image sensor 224. Even in a single lens microscope such as the imaging system of FIG. 2, if the dichroic mirror 218 is used instead of the bandpass filter 220, this epi-illumination method may be applied. When the excitation light 210 is irradiated to the sample placed on the sample stage 212, fluorescence is generated in the sample. The objective lens 214 collects the fluorescent light 216 from the sample and makes it parallel, and sends it to the tube lens 222. The role of the dichroic mirror 218 is the same as that of the bandpass filter 220 of FIG. 2. The fluorescence passing through the dichroic mirror 218 passes through the tube lens 220 and forms an image in the image sensor 224. In the imaging system of FIG. 3, the distance between the end face 212 of the sample to be imaged and the objective lens 214 is Z ₁ , the focal length of the objective lens 214 is f ₁ , the tube lens 222 and the image sensor 224. If the distance between Z ₂ and the focal length of the tube lens 222 is f ₂ , the cross section of the sample to be imaged and the focal point of the objective lens 214 irrespective of the distance between the objective lens 214 and the tube lens 222. When the distances are the same (Z ₁ = f ₁ ), the distance between the tube lens 222 and the image sensor 224 coincides with the focal length of the tube lens 222. That is, when Z ₁ = f _1, Z ₂ = f ₂ is established. At this time, the magnification M between the two-dimensional image size and the actual sample cross-sectional size measured by the image sensor 224 is expressed by Equation 3 below.

This infinity-corrected microscope system has the advantage of being able to sandwich various optical components, such as dichroic mirrors 218, between the objective and tube lenses of the image. Consists of. In an infinity-corrected system, the distance between the objective lens 214 and the tube lens 222 is not important, but when the magnification is large, the objective lens and the tube lens are reduced in order to reduce the so-called 'vignetting effect', which reduces the brightness of a part far from the center of the image. The distance is usually smaller than the sum of the focal length of the objective lens and the focal length of the tube lens.

In the optical imaging system using the (two-dimensional) image sensor 224 used in FIG. 2 or FIG. 3, only a specific cross section of the sample is focused on the image sensor 224, and light from another cross section is out of focus. . Therefore, the fluorescent light generated from the phosphor distributed on a specific cross section of the focused sample is distributed in a narrow area of the image sensor 224 and is measured at a high intensity, and in the phosphor distributed on another cross section of the non-focused sample. The generated fluorescent light is distributed in a wide area on the image sensor 224 and is weakly observed.

At this time, if the sample is scanned in the depth (z) direction by using the fixed image sensor 224 and a plurality of two-dimensional images are acquired in series, each image focuses only on fluorescent light corresponding to a specific cross section of the sample. This condensation and light from the rest of the cross section is distributed over a large area of the image. By arranging a series of two-dimensional images obtained by scanning the depth direction of such a sample in sequence, it is possible to obtain a three-dimensional image of a sample that can grasp the three-dimensional structure of the sample, although not complete.

In order to obtain a perfect three-dimensional shape of the sample, it must be able to completely block the light generated from the cross-section of the out-of-focus sample and distributed over a large area of the image sensor. In confocal microscopy, a physical method is used to pass fluorescent light generated from a sample through a pinhole, thereby obtaining a certain effect of blocking fluorescent light generated from an unfocused cross section. However, the present invention proposes a method of blocking fluorescent light generated at an unfocused cross section using a simple image processing method in software without using a physical method.

The light generated from the focused section is focused on the image sensor 224 to form a clear and distinct image, but the light generated from the unfocused cross section of the sample that reaches the image sensor is not focused and thus is not imaged. The sensor 224 has a characteristic of spreading spatially widely. Thus, passing a spatial high-pass filter, which passes only the high frequencies in the spatial frequency inverse, can easily remove most of the fluorescent light generated from unfocused sections. . By applying various algorithms to reduce spatial low frequency components or amplify spatial high frequency components in the measured image, it is possible to effectively remove the light generated from the unfocused cross-sections in various ways. You can get a new two-dimensional image composed of only the light. By moving the sample or image sensor 224 in the depth direction and measuring a two-dimensional image in series, two-dimensional images of light generated at various depths of the sample, that is, tomographic images of the sample, can be obtained. The dimensional shape may be generated (or may be referred to as 'reconstruction' or 'restore' herein).

Hereinafter, a method of obtaining a scanning image in a depth direction will be described in more detail.

Scanning in the depth direction (z direction) of the sample is necessary for three-dimensional reconstruction of the image. 2 or 3, several three-dimensional scan methods are applicable to these fluorescence microscope systems.

First, referring to FIG. 2 or 3, one of three-dimensional scanning methods is a method of scanning by moving the sample stage 212 in a depth direction (z direction) (hereinafter, referred to as “method A”). In this case, since the moving distance of the sample stage 212 and the moving distance in the depth direction of the measured three-dimensional image correspond one-to-one, the restoration of the three-dimensional image of the sample is simple.

2 or 3, there is a method of scanning by moving the image sensor 224 in the z-direction (hereinafter, referred to as “method B”) in the three-dimensional scanning method. In this case, there is a disadvantage that the moving distance of the sample stage 212 and the moving distance in the depth direction of the measured 3D image do not linearly correspond.

In addition, in the case of FIG. 3, there is a method of scanning by moving the tube lens 222 in the z direction (hereinafter, referred to as “method C”). In this case, the same result as the scanning method of the image sensor described in 'Method B' can be obtained. In this case, as in the above case, the moving distance of the sample stage 212 and the moving distance in the depth direction of the measured 3D image do not linearly correspond.

In addition, referring to FIG. 3, there is a method of scanning by moving the objective lens 214 in the z-direction among the three-dimensional scanning methods (hereinafter, referred to as 'method D'). In this case, in FIG. 2, the moving distance of the sample stage 212 does not linearly correspond to the moving distance in the depth direction of the measured 3D image. In the case of FIG. 3, since the moving distance of the sample stage 212 and the moving distance in the depth direction of the measured 3D image correspond one-to-one, restoration of the 3D image of the sample is simple.

In addition, in the case of FIG. 2 or 3, a method of scanning by using an additional zoom lens in front of the image sensor 224 and adjusting the position of the zoom lens (hereinafter, referred to as “method E”) is also possible.

In general, an image of a sample formed on the image sensor 224 by the objective lens 214 forms an image magnified M times than the sample. Since the magnification M usually has a magnification of 50 to 100 times, using a simple lens formula, the effect obtained by moving the sample by a small distance δZ is the same as that caused by moving the image sensor by -M ² δZ. Therefore, assuming an imaging system with M = 70, images obtained by scanning a sample every 100 nm can be obtained while scanning by 0.49 mm using Method (B) or Method (C), which is a method of scanning an image sensor or a tube lens. Can be. Therefore, scanning the image sensor or the tube lens has the advantage that the information of the depth direction of the sample can be obtained very accurately even by using the less accurate scanning method.

Hereinafter, an algorithm (s) for processing a plurality of two-dimensional images (of course, tomographic images having different depth values) and generating (reconstructing and reconstructing) a more precise three-dimensional image using the processed two-dimensional images ) Will be described in detail.

In an optical fluorescence imaging system using the two-dimensional image sensor 224, the application of various image processing algorithms can very effectively reduce the effect on fluorescent light generated at the cross-section of an out of focus sample. By effectively removing the light generated from the unfocused cross section, a two-dimensional tomographic image of a new sample can be obtained, consisting only of the fluorescent light mainly generated from the relatively focused section of the sample. The new two-dimensional tomographic images obtained by this method have characteristics very similar to those obtained from conventional confocal microscopes. The basic principle of the algorithm used here is to reduce the spatial low frequency component or increase the spatial high frequency component in the measured two-dimensional image. It is to enhance the intensity of the fluorescent light generated at a specific tomography surface and to reduce the fluorescent light generated at the other cross section.

First, if the i-th 2D image of the N-dimensional 2D images obtained from the series is scanned as I _i (x, y) while scanning the sample by a constant distance ΔZ in the depth direction, the i-th 2D image is used as a constant ΔZ in the depth direction Z-axis. A three-dimensional image I (x, y, z) sampled for each distance can be constructed. In this case, the sampling distance ΔX or ΔY of the X or Y-axis is the pixel-to-pixel spacing of the used two-dimensional image sensor divided by the magnification M of the imaging system to be used. ΔY = 200 nm. When the image is acquired while the sample is scanned in the depth direction by ΔZ, the sampling distance with respect to the Z axis is equal to ΔZ. When the image is acquired while scanning the objective lens 214, the tube lens 222, or the image sensor 224 at regular intervals, the sampling distance on the actual sample between the images is not the same. The scan distance of the objective lens 214, the tube lens 222 or the image sensor 224 should be scanned non-linearly. In the case of obtaining a two-dimensional tomographic image series by scanning the objective lens 214, the tube lens 222, or the image sensor 224 at regular intervals, the interpolation method and the lens formula between the images obtained for each point on the two-dimensional coordinates Equation 1 can be used to generate new series of tomography images with constant scan distance on the sample.

In order to process the measured three-dimensional image in the spatial frequency band, it is necessary to convert the original image to three-dimensional Fourier transform. Between the original image I (x, y, z) and the Fourier transform function F (f _x , f _y , f _z ), the relationship between the Fourier transform and the inverse transform, as shown in Equations 4 and 5, is satisfied.

Fourier Transform

Fourier Inverse

Here, f _x , f _y and f _z denote spatial frequencies corresponding to the X, Y and Z axes of the 3D image, respectively.

The equations shown in the following algorithms are based on a spatial low frequency component (F (f _x , f _y , f _z )) of the measured three-dimensional image I (x, y, z) and the three-dimensional Fourier transform Deform using a method to reduce the spatial low frequency component or increase the spatial high frequency component and then use the new three-dimensional cross-sectional image I _a (x, y, z) or its corresponding Fourier How to construct the converted new function F _a (f _x , f _y , f _z ).

First, the first algorithm I _a (x, y, z) = (I (x, y, z)) ² will be described as a possible algorithm.

Where I (x, y, z) represents the three-dimensional intensity of the image obtained by scanning the sample by the method proposed in the present invention, and I _a (x, y, z) applies the first algorithm. Reconstructed three-dimensional image. The function of Fourier transforming _a new image I _a (x, y, z) obtained by Algorithm 1 is called F _a (f _x , f _y , f _z ), which is a convolution theorem commonly used in Fourier transforms. The relationship shown in Equation 6 is satisfied.

는 수학적으로 convolution 연산을 나타낸다. 측정된 본래의 3차원 이미지 I(x,y,z)에 대한 푸리에 변환 함수 F(f_x,f_y,f_z)는 보통 공간주파수 f_x , f_y 및 f_z의 원점 부근에 피크가 있고 f_x , f_y 및 f_z의 크기가 커짐에 따라 값이 작아지는 형태를 갖는다. 이 함수의 자체 convolution 결과로 얻어지는 새로운 함수 F_a(f_x,f_y,f_z)는 공간주파수 f_x , f_y 및 f_z에 대하여 본래의 함수 F(f_x,f_y,f_z)보다 부드러운 함수가 된다. 즉, 새로운 공간주파수 함수 F_a(f_x,f_y,f_z)는 공간주파수의 원점부근에 있는 저주파 성분 피크는 줄어들고 상대적으로 고주파성분 성분은 비율은 원래의 함수 보다 커지게 된다. 따라서, I_a(x,y,z)는 측정된 3차원 이미지 I(x,y,z)에 서 샘플의 특정 단층면에서 발생된 형광 빛의 공간적 고주파 성분의 세기는 강화 시키고 나머지 단면에서 발생된 형광 빛의 공간적 저주파성분은 감소시킨 새로운 3차원 단층이미지이다. 이러한 새로운 3차원 단층이미지 I_a(x,y,z)는 측정된 원래의 이미지보다 샘플의 3차원적인 구조를 더욱 잘 나타내 주며 이는 기존의 공초점현미경을 통해서 얻을 수 있는 샘플의 3차원 구조 형광 이미지와 매우 유사하게 된다.here

Mathematically represents a convolution operation. The Fourier transform function F (f _x , f _y , f _z ) for the measured original three-dimensional image I (x, y, z) usually has a peak near the origin of the spatial frequencies f _x , f _y and f _z As the sizes of f _x , f _y, and f _z increase, the value decreases. Its new function obtained by the convolution result of this function than F _a (f _x, f _y, f _z) is the spatial frequency f _x, f _y, and the original function F with respect to _{f z (f x, f y} , f z) Becomes a smooth function That is, the new spatial frequency function F _a (f _x , f _y , f _z ) reduces the peaks of the low frequency components near the origin of the spatial frequency, and the ratio of the high frequency components becomes larger than the original function. Therefore, I _a (x, y, z) increases the intensity of the spatial high-frequency components of the fluorescent light generated at a specific tomographic plane of the sample in the measured three-dimensional image I (x, y, z) The spatial low frequency component of fluorescent light is a new three-dimensional tomographic image reduced. This new three-dimensional tomographic image I _a (x, y, z) represents the three-dimensional structure of the sample better than the original image measured, which indicates that the three-dimensional fluorescence of the sample can be obtained through conventional confocal microscopy. Very similar to the image.

Hereinafter, as another possible algorithm in addition to the first algorithm, the second algorithm F _a (f _x , f _y , f _z ) = G (f _x , f _y , f _z ) F (f _x , f _y , f _z ) ).

Here, G (f _x , f _y , f _z ) has a function of reducing the low frequency component or amplifying the high frequency component in the function F (f _x , f _y , f _z ) of the frequency component obtained by Fourier transforming the original image. Represents a typical high pass filtering function. In this case, a new image I _a (x, y, z) can be obtained by Fourier inverse transforming the newly obtained F _a (f _x , f _y , f _z ) using Equation 5. In general, G (f _x , f _y , f _z ) may be various functions as shown in Equations 7 to 13.

Where α and a are each positive integers (α> 0, a> 0). When the sampling distances for the X, Y, and Z axes are ΔX, ΔY, and ΔZ, respectively, the maximum spatial frequencies f _Mx , f _My , and f _Mz corresponding to each axis are 1 / ( 2ΔX), 1 / (2ΔY), and 1 / (2ΔZ).

The function F (f _x , f _y , f _z ) of the frequency component obtained by Fourier transforming the original image by multiplying the filtering function G (f _x , f _y , f _z ) by reducing the low frequency component or increasing the high frequency component The new spatial frequency function F _a (f _x , f _y , f _z ) obtained reduces the peaks of the low frequency components near the origin of the spatial frequency and the ratio of the high frequency components becomes larger than the original function. Thus, _a new image I _a (x, y, z) obtained by Fourier inverse transforming F _a (f _x , f _y , f _z ) occurs at a particular tomographic plane of the sample than the measured three-dimensional image I (x, y, z) It is a new three-dimensional tomographic image that enhances the intensity of the spatial high frequency components of the fluorescent light and decreases the spatial low frequency components of the fluorescent light generated in the remaining sections. This new three-dimensional tomographic image I _a (x, y, z) represents the three-dimensional structure of the sample better than the original image measured, which indicates that the three-dimensional fluorescence of the sample can be obtained through conventional confocal microscopy. Very similar to the image.

4 is a flowchart illustrating a high speed imaging method according to at least one embodiment of the present invention. Referring to Figure 1 as follows.

The imaging unit 110 scans a sample in a depth direction and acquires two-dimensional images by capturing a plurality of times (operation 410).

The image reconstructor 120 generates a 3D image of the sample (named 'reconstruction' or 'restore') using the 2D images acquired in step 410 (step 420).

The program for executing the high speed imaging method according to the present invention mentioned above on a computer may be stored in a computer readable recording medium.

Here, the computer-readable recording medium may be a magnetic storage medium (for example, a ROM, a floppy disk, a hard disk, etc.), and an optical reading medium (for example, a CD-ROM, a DVD). : Digital Versatile Disc).

So far, the present invention has been described with reference to the preferred embodiments. Those skilled in the art will appreciate that the present invention can be implemented in a modified form without departing from the essential features of the present invention. Therefore, the disclosed embodiments should be considered in descriptive sense only and not for purposes of limitation. The scope of the present invention is shown in the claims rather than the foregoing description, and all differences within the scope will be construed as being included in the present invention.

1 is a block diagram illustrating a high speed imaging apparatus according to at least one embodiment of the present invention.

2 is a reference diagram of a system representing a high speed imaging apparatus according to a first embodiment of the present invention.

3 is a reference diagram of a system representing a high speed imaging apparatus according to a second embodiment of the present invention.

4 is a flowchart illustrating a high speed imaging method according to at least one embodiment of the present invention.

Claims (17)

An imaging unit which scans a sample in a depth direction and acquires two-dimensional images by capturing a plurality of times; And An image reconstruction unit generating a 3D image of the sample using the 2D images; Wherein the image reconstruction unit squares the three-dimensional image to reconstruct a new three-dimensional image. The method according to claim 1, And the imaging unit acquires the two-dimensional images by moving the sample in a depth direction and photographing the sample a plurality of times. The method according to claim 1, And the imaging unit moves the imaging device in the imaging unit in a depth direction to acquire the two-dimensional images by photographing the sample a plurality of times. The method according to claim 1, And the imaging unit moves the tube lens in the imaging unit in a depth direction and acquires the two-dimensional images by photographing the sample a plurality of times. The method according to claim 1, And the imaging unit moves the objective lens in the imaging unit in a depth direction and acquires the two-dimensional images by photographing the sample a plurality of times. delete delete An imaging unit which scans a sample in a depth direction and acquires two-dimensional images by capturing a plurality of times; And An image reconstruction unit generating a 3D image of the sample using the 2D images; Wherein the image reconstruction unit obtains a spatial frequency function by multiplying a function of a frequency component obtained by Fourier transforming the 3D image by a filtering function that reduces a low frequency component or increases a high frequency component, and inverses the obtained spatial frequency function. A high speed imaging device comprising Fourier transform to reconstruct a new three-dimensional image. (a) scanning a sample in a depth direction and capturing a plurality of times to obtain two-dimensional images; And (b) generating a 3D image of the sample using the 2D images; Wherein, step (b) is a high-speed imaging method, characterized in that the square of the three-dimensional image to reconstruct a new three-dimensional image. The method of claim 9, In the step (a), the 2D images are acquired by moving the sample in the depth direction and capturing the sample a plurality of times. The method of claim 9, In the step (a), moving the image pickup device in the image pickup unit in the depth direction to obtain the two-dimensional images by taking the sample a plurality of times. The method of claim 9, The step (a) is a high-speed imaging method of acquiring the two-dimensional images by moving the tube lens in the imaging unit in the depth direction and taking the sample a plurality of times. The method of claim 9, The step (a) is a high-speed imaging method of obtaining the two-dimensional images by moving the objective lens in the image pickup unit in the depth direction to take the sample a plurality of times. delete delete (a) scanning a sample in a depth direction and capturing a plurality of times to obtain two-dimensional images; And (b) generating a 3D image of the sample using the 2D images; Wherein the step (b) includes obtaining a spatial frequency function by multiplying a function of frequency components obtained by Fourier transforming the three-dimensional image by a filtering function for reducing low frequency components or increasing high frequency components, and obtaining the spatial frequency functions. The high-speed imaging method of claim 4, wherein the inverse Fourier transform is used to reconstruct a new three-dimensional image. A computer-readable recording medium storing a computer program for executing the method of any one of claims 9 to 13 and 16 on a computer.

Images